Production Optimization of an Existing District Heating Network with Multiple Heat Producers

A Mixed Integer Linear Programming (MILP) optimization problem is formulated to optimize the power production of several heat production units and a storage tank in a district heating network in Weil am Rhein, Germany. This implementation is the first step in the development of a Model Predictive Control (MPC) methodology in the research project WOpS to operate networks with decentralized heat producers. Historical demand data and producer specifications of the existing network are used for the analysis. The operation is simulated for a winter and a summer day. It is shown that the methodology is successfully implemented, even though more constraints should be considered in future developments to include dynamic effects more realistically.